© Copyright 2026 Mastersinlibraryscience.org.
How MEDLINE reevaluation works, how to verify indexing status, and what to tell patrons and faculty
What happens when PubMed puts a journal on hold?
On October 20, 2026, Rafal Marszalek, chief editor of Scientific Reports, confirmed in an email seen by Retraction Watch that the National Library of Medicine had paused MEDLINE indexing and PubMed Central archiving of the journal's newly published content while it reviews the title. Scientific Reports is the largest journal in PubMed by indexed article count, so the pause is a bibliographic-control event, not a routine queue delay.
For librarians, the work begins where the news stops: confirming what is actually frozen, checking the discovery layers that assume coverage, and answering faculty who treat "indexed" as a permanent property of a title. It never was.
A journal can look fully active and still be frozen out of the biomedical discovery pipeline. A platform outage, an ordinary indexing queue, and a formal NLM reevaluation pause are three different events with three different consequences. For librarians, the practical question is which of these three states a journal is actually in.
An outage is technical and temporary: PubMed or PMC goes down, then comes back. Ordinary indexing lag is the normal wait while citations move from in-process to indexed for MEDLINE. Most citations make that progression, though not all will. A hold is an NLM action about review status. It is not a retraction, a publisher withdrawal, or a finding that the science is fraudulent.
In the Scientific Reports case, NLM told the journal that new MEDLINE indexing and new PMC archiving were paused, so new content would not appear in PubMed. The pause is prospective. Older records stay searchable, the journal keeps publishing, and readers can still reach articles through the publisher.
NLM Catalog uses a field labeled "Current Indexing Status." A journal in good standing reads "Currently indexed for MEDLINE," with an "In" field showing the indexed period. A title under review may read "Not currently indexed." That means the journal is not presently receiving MEDLINE indexing. It does not mean the journal was never indexed, and it does not erase earlier citations.
The consequence is narrow but serious: the journal still publishes and authors still cite it, while the discovery pipeline stops accepting new deposits.
Librarians and patrons often use MEDLINE, PubMed, and PubMed Central as if they were one thing. They are three separate systems with separate rules, and a pause applied at one layer does not automatically stop the others. The table below sets out what each one holds and what a MEDLINE indexing hold actually touches.
| System | What it contains | What a MEDLINE indexing pause changes | How records get there |
|---|---|---|---|
| MEDLINE | Bibliographic citations to journals selected for MEDLINE, indexed with NLM Medical Subject Headings. | Fully indexed citations are held before release. Indexing staff may keep working, and in-process citations can still appear. | NLM indexers complete each citation and the record is added to PubMed on the normal update schedule. |
| PubMed | Citation information only: titles, authors, journal, publication dates, and abstracts. Full text is not part of a PubMed record. | Only fully indexed MEDLINE citations are held back. Publisher-supplied and in-process citations continue to be added. | Several routes feed PubMed, including publisher submissions and the citation flow from PMC. The MEDLINE filter narrows results to indexed citations. |
| PubMed Central (PMC) | A full-text archive of deposited final articles from participating journals, eligible collections, and funder-supported research. | Not automatically affected. PMC deposit follows journal, publisher, or author-manuscript arrangements and PMC technical requirements. | Publishers or authors deposit articles. For full-participation or NIH portfolio journals, PMC sends the citation to PubMed once the article is fully processed. |
NLM does not treat indexing as a one-time stamp. Selection and continued participation rest on the same evidence base, so a librarian who knows it can explain a pause without guessing.
Three inputs carry the decision: scientific policy set by the NLM Board of Regents, the journal's suitability for the NLM Collection under the Collection Development Guidelines, and recommendations from external expert consultants. No single reviewer settles it. Multiple people inside and outside NLM assess each journal, and the outcome rests on their combined judgment.
At least two consultants generally review each journal, usually one doctoral-level scientist or physician and one medical librarian. They work through roughly the last two years of published articles, reading primarily for scientific and editorial quality.
Journals already in MEDLINE must report changes to title, ISSN, ownership, publisher, or publishing format, along with mergers, and must flag significant shifts in scope, editorial board, or editorial policies. Reevaluation also opens when problems in published articles go unaddressed or appear systemic, or when verifiable concerns about scientific or editorial quality surface.
When a significant change triggers review, NLM waits until either 20 primary, peer-reviewed articles have appeared since the change or one year has passed, provided enough primary content exists. The policy sets that content threshold rather than a fixed calendar length, so a hold can run longer than faculty expect.
The reevaluation then follows the path of a new application, external consultants included. The journal re-earns its place rather than coasting on history, the framing to give a department wondering why a familiar title suddenly stopped appearing.
The Scientific Reports pause shows that scale in a biomedical journal is not the same thing as a settled relationship with the National Library of Medicine.
The dates matter. Rafal Marszalek, the journal's editor in chief, communicated the review to the journal's readership in an October 20, 2026 email, and Retraction Watch reported the pause on October 30, 2026. During the review, MEDLINE indexing and PubMed Central archiving of newly published content are suspended, which means new articles do not appear in PubMed by either route. The publisher is Springer Nature.
Marszalek cited the journal's "significant growth in recent years" as context. He did not state why NLM opened the review, and no public NLM statement has supplied a reason. That gap is itself part of the lesson: reevaluation is an internal collection-development judgment, and journals learn the outcome, not always the analysis behind it.
Retraction Watch described Scientific Reports as the largest journal in PubMed by articles indexed. A title can rank first in volume and still find its vetting under review. For collection development, track them separately: citation counts and output measure reach, while active MEDLINE standing measures a continuing relationship with the indexing body.
As of November 4, 2026, the review remained ongoing. No final NLM decision, reinstatement date, or permanent delisting had been announced. Marszalek said the journal was in contact with NLM and expected indexing to resume, including for content published during the pause. That is the editor's expectation, not an announced outcome, and libraries should describe it that way.
Indexing status rarely breaks anything loudly. It breaks a chain quietly, one link at a time.
A licensed full-text package does not vanish because indexing paused. Resolvers generally model discovery coverage separately from institutional access, so old links keep working while new articles never arrive.1 The title still sits in your A-Z list with the same dates, because nobody edited it. That silence is the real problem: nothing looks broken, and the gap surfaces months later when a patron cannot find a 2026 paper that plainly exists.
Search dates carry more weight than usual. Report the date you searched, name the journal whose current content is not yet indexed, and describe how you checked. Note whether the pause began before or during your search window. Hand-search the journal's own site for that period and log the step in your flow diagram. Citation tracking is affected too: a record absent from PubMed is a record other authors may never cite.
Flag the title in your ERM, add an internal note with the date you verified the hold, and set a recheck cadence rather than waiting for a vendor alert that may never arrive.3
NLM's 2024 List of Serials Indexed for Online Users runs to 15,473 serial titles; only 5,294 carry current MEDLINE indexing. That gap is the reason this check belongs in a written procedure rather than a glance at a badge on a publisher's homepage.
Search the NLM Catalog by ISSN first, falling back to exact title. Read the field labeled Current Indexing Status for the exact phrase "Currently indexed for MEDLINE." The coverage field, labeled In, shows the volumes or dates covered, which matters because a journal can be current while indexing begins only from a stated year or volume. Use the currently indexed filter in the results sidebar as a second check. Presence in a catalog result alone proves nothing: hits may reflect previous titles, journals no longer indexed, or citations surfacing through another NCBI database.
MEDLINE and PMC are separate databases, so verify each one on its own. The PMC Journals list at ncbi.nlm.nih.gov/pmc/journals is searched by title or ISSN. Free articles on a journal site, or a PubMed record for a single older paper, establish neither. A PMC listing also does not confirm that every volume has been deposited.
Compare the date of the most recent indexed record with the journal's publication schedule. Routine lag fits only when the catalog still reads "Currently indexed for MEDLINE," coverage dates remain, and no NLM notice or record identifies a hold. Newly selected journals typically begin with the current year or volume, so a timing gap is normal. NLM publishes no universal day or week cutoff, which makes the comparison journal-specific. A hold requires explicit NLM wording; silence is not one.
Record the date you checked, the page you used, and the exact status wording. Promotion packets and collection development files both need a snapshot someone can cite months later.
Indexing status functions as a quality endorsement that libraries never formally issued: when we build discovery around PubMed coverage, we inherit a judgment NLM made and repeat it every time a patron clicks through.
Reference-desk conversations about journal quality have shifted from a yes-or-no indexing check to a discussion of status over time, and the NLM review of a high-volume title shows why.
Liaison librarians need a sentence they can repeat without hedging: indexing status is one signal among several, and a hold means the journal is under review, not discredited. An NLM reevaluation is a formal status action. It is not a predatory-publishing finding, and saying so plainly keeps a temporary pause from being retold as an accusation.
For faculty assembling a dossier, a single "is it indexed?" check is fragile. Help them document which indexes cover the journal and when. Record the coverage dates, whether the record shows MEDLINE indexing, PubMed visibility through another route, or PMC participation1, and the date the status was verified. A dated note survives a status change; a memory does not.
Keep the categories distinct. A legitimate high-volume journal under review typically has an identifiable editorial board, stated fees and waivers, honest peer-review claims, and a real publication history.3 Predatory indicators look different: fabricated indexing claims, editors who cannot be verified, concealed fees, false peer-review promises, aggressive solicitation.2
Pair indexing with complementary evidence, including DOAJ listing for open access titles, Scopus or Web of Science coverage, board transparency, and retraction and correction patterns.3 A new journal may simply have no indexing yet, and absence alone proves nothing.
A journal that loses MEDLINE indexing over scientific-quality concerns cannot reapply for two years after that review concludes, and the waiting period is one of the few parts of the process NLM states plainly.
NLM support guidance indicates that citations already indexed are retained, subject to exceptions the agency does not itemize publicly. The NLM Catalog keeps current indexing status separate from the year a journal was indexed from, so the earlier coverage window stays visible in the record even after status changes. In practice, a deselected journal's backfile does not disappear from PubMed.
NLM has published neither a blanket commitment to back-index everything that appeared while a journal was on hold nor an explicit exclusion. Corrections and retractions attached to already-indexed articles can receive a case-by-case exception, but routine new articles fall outside that allowance. Librarians should describe the outcome as undetermined rather than predict it in either direction.
Publishers apply through the MEDLINE Publisher Portal. If no application arrives within 90 days of notification, NLM stops indexing and bars further submissions. Appeals are decided within 30 calendar days, and the outcome is final. An appeal succeeds only where a policy requirement was misapplied, not where a journal disputes the scientific judgment. A successful appeal triggers re-review of the original application with no new materials; a denied one leaves the original reapplication date standing.
Deselection is uncommon and usually quiet, typically following very late publication, sweeping editorial change, or a publisher transfer. Full retroactive back-indexing after a hold is essentially undocumented in the public record, which is why most libraries have never written a procedure for it.
